© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Joseph Bloom · Open Source Sparse Autoencoders for all Residual Stream Layers of GPT2-Small
    5. GPT2-Small
    7. Residual Stream
    9. 8-RES-JB
    11. 8063
    Prev
    Next
    INDEX
    Explanations

    proper nouns or names of people, places, or organizations

    oai_token-act-pair · gpt-3.5-turbo

    prominent names and entities in various contexts, including politics, sports, and business

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 8-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.8.hook_resid_pre
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    Skylion007/openwebtext
    Features
    24,576
    Data Type
    torch.float32
    Hook Point
    blocks.8.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    8
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    é¾įå
    -0.75
    âĹ¼
    -0.73
    }.
    -0.72
    Els
    -0.68
    âĶĢâĶĢâĶĢâĶĢ
    -0.68
     Azerb
    -0.66
    _.
    -0.65
    Ô
    -0.63
    Translation
    -0.61
    ãĢĤ
    -0.61
    POSITIVE LOGITS
     awoke
    0.77
     celebrates
    0.77
     unveiled
    0.73
     prepares
    0.72
     announced
    0.72
     announces
    0.71
     enters
    0.70
     apologised
    0.70
     reacts
    0.69
     reportedly
    0.68
    Activations Density 0.533%

    No Known Activations